Genetically modified mice, modified and engrafted mice, and their use in modeling human diseases, e.g., for the purpose of drug testing, are known in the art. Attempts have been made to use genetically modified mice to model a human immune system. A review of that field is provided in Manz (2007) Human-Hemato-Lympoid-System Mice: Opportunities and Challenges, Immunity, 26:537-541 hereby incorporated by reference.
To date no genetically modified mice have been generated that demonstrate infectivity with certain human pathogens, e.g., Salmonella typhi (S. typhi). Even for pathogenic infections for which mouse models exist, the models can fail to adequately model certain pathologies in humans, e.g., failure to form well-defined granulomas or granulomas containing human immune cells in mouse models of Mycobacterium tuberculosis (M. tuberculosis). In order to study the effects of certain pathogens on humans, and to test drugs for effectiveness in treating humans infected with certain pathogens, it would be useful to have a non-human animal such as a mouse that is genetically modified so that it is susceptible to infection with such a pathogen, e.g., S. typhi, and/or that the infection more closely models human pathology, e.g., more closely models a human infection of M. tuberculosis. 
In general, there is a need for genetically modified mice that can support maintenance and propagation of human hematopoietic stem cells, and for mice suitable for engraftment that can model or approximate certain aspects of a human hemato-lymphoid system, e.g., in response to a human pathogen.